Deakin Research Online This is the published version: Peszynski, K., Corbitt, B. and Coulthard, Darryl 2005, The politics of systems implementation: a critical approach, in Proceedings of the qualitative research in IT & IT in qualitative research, Griffith University, Institute for Integrated and Intelligent Systems, Brisbane, Qld. Available from Deakin Research Online: http://hdl.handle.net/10536/DRO/DU:30005741 Every reasonable effort has been made to ensure that permission has been obtained for items included in Deakin Research Online. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au Copyright: 2005, Griffith University, Institute for Integrated and Intelligent Systems.

The Politics of Systems Implementation: A Critical Approach

Abstract

*Konrad J. Peszynski **Brian Corbitt

*Darryl Coulthard Deakin University

* School of Information Systems Burwood, Victoria, Australia

**Graduate School of Management, Shinawatra University,

Bangkok 10900, Thailand; Email: konradp@deakin.edu.au Email: bcorbitt@shinawatra.ac.th

Email: dwcoulth@deakin.edu.au

The central concern of this paper is to identify the role of power and politics in systems implementation. The current literature misses the complexities involved in systems implementation essentially de-contextualising the implementation process. A post-structuralist view of power as both an obvious and hidden concept has provided the researcher a lens through which the selection and implementation of an enterprise-wide learning management system can be observed. This approach has important implications for research methods as the qualitative approach needed is challenged by acceptance of bias and the need to expose it as an important factor in explaining success or failure in systems implementation.

Keywords: System selection and implementation, power, politics, critical epistemology, social dramas

INTRODUCTION

The central concern of this paper is to identify the role of power and politics in systems implementation. The current literature on systems implementation is typically divided into two areas, process modelling and factor based studies. Process modelling classifies the implementation into a linear process, whereas factor based studies have argued that in order to "successfully" implement a system, particular critical factors are required. This literature misses the complexities involved in systems implementation through the human factors and political nature of systems implementation and is simplistic in its nature and essentially de-contextualises the implementation process.

The failure of an information system in an organisation can have a crippling effect on the organisation itself, the members of that organisation and the reputation of that organisation. In other words, if a system is not successfully implemented, the company could lose potential profit. For example, RMIT, a University in Victoria, Australia, mismanaged the implementation of the Academic Management System, resulting in students not being able to arrange enrolments, invoices, subject changes and timetables (Royall, 2002). Subsequently, the system was abandoned and members of the senior management and project team were fired.

Stories of failed systems occur regularly in the media, which damage the credibility of organisations and turn potential customers away. The costs of systems failure vary. Typically they include firstly, economic costs, such as investments in equipment and labour. Secondly, there are costs of missed opportunities, where a system fails to deliver on benefits promised. Finally, there are costs incurred in terms of client service or risks to the community (Sauer, 1993). In the case of the failed implementation at RMIT, the users of the system, students were unhappy with the system and called for the resignation of the Chancellor of the University, followed by the Vice-Chancellor (Royall, 2002). There are many other stories of failed systems implementation that have also had significant consequences. Neumann (1993, p. 146) highlights fourteen different cases of failed systems implementation. Examples include:

A new child support checking system in Virginia, USA, which experienced massive delays, confusion, lost checks, delayed payments, and improper seizure of tax refunds. Operations costs were expected to be triple the original estimates;

The Bank of America spent US$23 million on an initial 5-year development of MasterNet, a new computerized trust accounting and reporting system. After abandoning the old system, they spent US$60 million more trying to make the new system work, and finally gave up. Departed customer accounts may have exceeded billions of dollars; &

• Oklahoma hired a major accounting firm in 1983 to design a US$.5 million system to handle its workers' compensation claims. Two years and more than $2M later, the system still didn't work. It was finally finished in 1987 for nearly US$4 million;

It is therefore important to implement successfully a system and ensure that it is running well and that the users, both customers and employees alike, are sufficiently satisfied with the system to want to continue using it. The information systems community need a deeper understanding of what is involved in systems implementation, and more specifically, what enables successful implementation and what contributes to unsuccessful systems implementation. The focus of this study is on the social aspects of system selection and implementation as opposed to the technical or actual system implementation process. The researcher is interested in how the social component of system selection and implementation affects the overall process.

SYSTEMS IMPLEMENTATION

The technical component of system selection and implementation has already received significant discussion in the literature. Two approaches to systems implementation, process models and factor-based studies have been discussed extensively in the literature.

Process Models

Proponents of the first approach, process models, argue that systems implementation should follow a particular procedure, typically a five or seven-step process known as the system development lifecycle (SDLC) (Davis, 1974; Avison and Fitzgerald, 2003). This approach has created multiple methodologies, or stock or package of methods, such as the Structured Analysis, Design and Implementation of Information Systems (STRADIS) (Gane and Sarson, 1979); the Structured Systems Analysis and Design Methodology (SSADM) (Downs et al., 1988); and Object­oriented analysis (Coad and Yourdon, 1991). These methodologies can then be employed in the implementation process, allowing the implementation team to get various perspectives from the stakeholders of what they require, and how they can use that information to implement a system. This systems implementation approach then is essentially reductionist, perhaps simplistic. It is certainly structuralist. There is a constant search for categorisation, stages and sub-stages. To counter this apparent reliance on structure, some researchers have sought to understand those factors considered critical to the success of systems implementation (Rockart, 1979; Ginzberg 1981).

Critical Success Factors

Advocates of the second approach, factor-based studies, argue that in order to successfully implement a system, certain critical factors are required to be in place (Rockart, 1979; Ginzberg, 1981; DeLone and McLean, 1992, 2003). Typically, research has indicated that the absence of top management support (Ginzberg, 1981), poor attitudes towards information systems (Corbitt, 1997) and absence of education and training (Cragg and King, 1983) lead to failure of the information system implementation. Instead, the support of management, clear goals and objectives of the proposed system, project management and the available information technologies (Somers and Nelson, 2001) all contribute to a successful system. Implementation however, is neither driven entirely by factors of success or failure (Corbitt, 1997). Rather, the implementation process in information systems is reflective of the stakeholder relationship interactions and the impact of the context, either business, organisational, social or cultural, in which the implementation occurs.

The Importance of Power and Politics in Systems Implementation Research

It is argued that both of these approaches are structuralist, and essentially over simplistic in nature. These approaches do not provide adequate detail about the complexity of the process, nor do they tell us anything specific about the factors or steps involved with systems implementation. Such factor studies are rarely reflective of the processes which occur in systems implementation. Rather, the traditional approaches tend to provide structure to enable understanding of a complex process. They reduce the complex to an easier, simpler structure. Implementation is neither driven entirely by factors of success or failure (Corbitt, 1997). Rather, the implementation process in information systems is more reflective of the stakeholder relationship interactions and the impact of the context,

either business, organisational, social or cultural, in which the implementation occurs. However, the traditional approaches ignore or underplay the political aspects involved in stakeholder relationships, as well as elements of power in systems implementation.

As a result, some authors have taken a more socio-technical approach to information systems implementation (Orlikowski, 1992; Mitev, 2001). In order to do this, we as researchers, must "move beyond commonsense explanations of failure and success and find more complex and richer ways of understanding the use of IS in organisations through the inclusion of broader social, economic, political, cultural and historical factors" (Mitev, 2001, p. 84). By taking this approach, we can enable a better understanding of the power and politics involved in systems implementation, by focusing on social issues in the implementation process.

Rather than take the social aspect of systems implementation at face value, we need to understand and perform research that recognises the complexity and historical construction of the members of the implementation team and process (Mitev, 2001). The existing literature cannot describe or explain the political environment in systems implementation because politics in implementation endures influence, pressure, dogma, expediency, conflict compromise, intransigence, resistance, error, opposition and pragmatism (Ball, 1990). That is, the implementation process is complex, messy, inconsistent, ambiguous and contains dilemmas. As a result, the following research question is posed:

What is the role of power and politics in systems selection and implementation?

By undertaking this research, we can identify outcomes of the selection and implementation of an enterprise-wide system by providing a better understanding to the human factors involved, and specifically the power and politics in systems selection and implementation. It should be noted, however, that the outcomes of this study do not aim to provide a solution to the power and politics involved in systems selection and implementation, but to fully recognise that there are political factors involved in implementing a system.

It is argued in this paper that human factors in systems implementation are constantly changing and essentially operating in a dynamic relationship affecting the implementation process. The concept of power relations, as proposed by Foucault (1977, 1978, 1980), have been utilised in order to identify the dynamic nature of power and politics. Foucault (1978) argued that power is a dynamic set of relationships constantly changing from one point in time to the next. It is this recognition that is lacking from information systems. Furthermore, these power relations are created through the use of discourse. Discourse represents meaning and social relationships, forming both subjectivity and power relations. Discourses are also the practices of talk, text and argument that continuously form that which actors speak.

A post-structural view of power as both an obvious and hidden concept has provided the researcher a lens through which the selection and implementation of an enterprise-wide learning management system can be observed. The framework aimed to identify the obvious process of system selection implementation, and then deconstruct that process to expose the hegemonic nature of policy, the reproduction of organisational culture, the emancipation within discourse, and the nature of resistance and power relations. A critical case study of the selection and implementation of an enterprise-wide learning management system at the University of the Eastern Antipodes will be presented providing an in-depth investigation of the implementation of an enterprise-wide learning management system, spanning five years. This critical case study was analysed using social dramas to distinguish between the front stage issues of power and the hidden discourses underpinning the front stage dramas.

METHODOLOGY The objective of this research is to provide an understanding of the role of power and politics in systems implementation. As a qualitative approach has been adopted for this research, the researcher wishes to tell the story of the selection and implementation of an enterprise-wide learning management system at the University of the Eastern Antipodes. Pseudonyms have been provided for all people, places and other identifiable names in order to keep anonymity of members. Furthermore, the nature of this research suggests that a post-structuralist analysis of power and politics in systems selection and implementation is required in order to provide a richer understanding of the phenomena under investigation. A critical theory approach has been adopted in this study as the researcher is then able to critically assess the social reality being studied, and therefore create awareness and understanding of the various forms of social domination.

Critical Epistemology

Ngwenyama and Lee (1997, p. 153) have argued that the critical approach is different to the positivist, in that the critical approach, as adopted in the Information Systems discipline, observes people "not as passive receptacles of whatever data or information that is transported to them, but as intelligent actors who assess the truthfulness, completeness, sincerity, and contextuality of the messages they receive." The critical approach also differs from the interpretive approach, as the critical approach "requires the researcher to attend to not only the matter of mutual understanding, but also the matter of the emancipation of organisational actors from false or unwarranted beliefs, assumptions and constraints" (Ngwenyama and Lee, 1997, p. 153-154).

Data Collection Techniques

To facilitate clarity and to provide some boundaries for the study, a case study methodology has been adopted, as it allows the researcher to identify what it is the subjects are doing, in their own words. Benbasat et a!. (1987, p. 370) note that a case study allows for the exploration of "a phenomenon in its natural setting, employing multiple methods of data collection to gather information from one or a few entities." Benbasat et a!. (1987) argue that case study research is well-suited to the information systems discipline not only because "the researcher can study information systems in a natural setting, learn about the state of art, and generate theories from practice" (Benbasat et a!., p.81), but also "to understand the nature and complexity of the processes taking place" (Benbasat et a!., p.81). The benefit of using a case study in developing a deeper understanding is that its method usually focuses on one or a few cases to be investigated to represent other typical cases. A case study also covers a small scope of a case and investigates the case in detail (Denscombe, 1998). By being immersed in the organisational activities, the researcher is able to gain richer information as to how power relations are created and how they may transform over time for the systems implementation group.

Data was collected from interviews and existing organisational documents. Staff involved in the implementation of the learning management system in the University of the Eastern Antipodes were interviewed and their data supplemented by and cross referenced with university policy documents, minutes of meetings and other publications. The enterprise-wide learning management system was selected and implemented to enable academic staff to manage teaching and learning by keeping track of student progress and performance across all types of learning activities. They used Ii range of tools to create learning resources, deliver content, monitor student participation and assess student performance. The systems selection and implementation process studied in this paper has been on-going since 2000. The systems selection and implementation process of the enterprise-wide learning management system has been written in narrative form, telling the story of the selection and implementation process at the University of the Eastern Antipodes. This approach has been adopted as the narrative form "supports a unity of form among the original interview situation, the analysis, and the final report" (Kvale, 1996, p. 184).

A case study methodology was employed, thus enabling the researcher to get inside the organisation and conduct multiple interviews and document collection in order to increase the rigour and credibility of the research. This approach also enabled the researcher to describe the organisation from the members point-of-view, reducing the distance between the researcher and the members of the study. By being immersed in the organisational activities, the researcher was able to gain richer information and explore the power relations created in a systems selection and implementation group, and how these power relations transform over time. Sixteen members of the systems implementation team were involved in this research out of 23 potential members.

Data Analysis Techniques

The concept of social dramas was utilised as a tool for data analysis and was adopted from the work of Victor Turner in the Anthropological discipline. Social dramas maintain that a social system and the inherent social relations of people or groups establishes and re-establishes a social equilibrium (Turner, 1957). However, balances are typically disrupted through political, cultural and social challenges. As a result, social disequilibrium or imbalance occurs. Readjustments need to happen in order to restore the balance. As Turner (1957) argued, when readjustments are made "profound modifications" may occur. In other words, the new social equilibrium is rarely a replica of the old equilibrium. By viewing systems implementation as social drama, we are able to vividly observe how social movements operate in practice (Turner, 1980). According to Turner (1957, p. 161), "the interests of certain persons or groups may have gained at the expense of others ... certain relations between persons and groups may have increased in intensity, while others may have diminished." We are essentially seeing challenging and political movements inside the social system. People, or groups object to one another, initiating a social drama.

The concept of a social drama refers to a series of events in which there are shifts in power, views or opinions, and changes in social groups in which the social drama is operating (Turner, 1957, 1980; Corbitt, 1995, 1997). Social

dramas occur within groups of persons who "share values and interests and who have a real or alleged common history" (Turner, 1980, p. 149). As an idea is contested it leads to a challenging of what currently exists. The actors act out social dramas by developing their interpretation of what has happened and what should happen. Social dramas occur in a context that is defined geographically by location (arenas) and by the underlying values, social constructs and meanings attached to statements and action by each of the actors involved. In the implementation of a system there appears to be a series of events, contestations, struggles, discourse setting, crises or 'social dramas,' which the actors in the implementation process go through (Corbitt, 1997). It is argued that implementation is rarely an ordered or sequential process. Actors within implementation contest and reconstruct the system to achieve their goals, to maintain their ideologies, to change programs, to change existing ideologies or to shift real power.

The social drama concept has been adopted in this study as it allows the researcher to create a framework to view the political nature of the systems implementation process. By ultimately breaking down the public episodes of tensional irruption into smaller periods, the researcher can examine the power relations and political nature of that period through the eyes of the members involved in the study. By employing the technique of social dramas, the researcher can also note the transforming power relations, according to Foucault (1977; 1978; 1980) from period to period.

In conjunction with the social drama metaphor, discourse analysis was also employed to make sense of, and analyse the collected interviews and documents. According to Howarth and Stavrakakis (2000, p. 4), discourse analysis refers to "the practice of analysing empirical raw materials and information as discursive forms. This means that discourse analysts treat a wide range of linguistic and non-linguistic data - speeches, reports, manifestos, historical events, interviews, policies, ideas, even organisations and institutions - as 'texts' or 'writings'." In other words, discourse analysis puts words into work, giving them meaning, constructing perceptions and formulating understanding and ongoing courses of interaction (Gubrium and Holstein, 2000). By utilising discourse analysis as a method to analyse, it is noted that the researcher does not start discourse analysis from texts and interactions, but from social issues and problems, problems which face people in their social lives, issues, according to Fairclough (2001), which are taken up within sociology, political science and/or cultural studies. To perform discourse analysis, "empirical data are viewed as sets of signifying practices that constitute a 'discourse' and its 'reality', thus providing the conditions which enable subjects to experience the world of objects, words and practices" (Howarth and Stavrakakis, 2000, p. 4).

THE SELECTION AND IMPLEMENTATION OF AN ENTERPRISE-WIDE LEARNING MANAGEMENT SYSTEM The illustrating case study provided here is the selection and implementation of an enterprise-wide learning management system in the University of the Eastern Antipodes. This learning management system was implemented, according to members interviewed, in order for academic staff to purportedly manage learners, the students, by keeping track of their progress and performance across all types of training activities. The learning management system allows staff to create learning resources, deliver content, monitor student participation and assess student performance. This learning management system was implemented following a push by the senior members of the university to centralise the processes of the university and create a sense of uniformity across the various schools and faculties.

Phase One of Systems Implementation

A decision was made towards the mid- to late nineties by senior members of the university to centralise systems. Each faculty was operating their own learning management system, depending on what focus that faculty had - either on- or off-campus. This resulted in the selection of the enterprise-wide learning management system, EducateOnline, which, according to members, was poorly selected and rushed in. Some members mentioned that EducateOnline was selected in order to gain an international business contract. However, according to members interviewed, EducateOnline was the only system that fulfilled the Department of IT-led requirements of running on Oracle.

In 1999 a new Deputy Vice-Chancellor was appointed and pushed the corporate line of having one enterprise-wide learning management system that ran on Oracle. Members from the largest faculty in the university, the Faculty of Commerce and Administration, and users of another learning management system, CommunicateOnline, criticised this decision, but the Deputy Vice-Chancellor kept pushing for a centralised system. After EducateOnline ran for one year, it was discovered that they were moving out of the higher education market and into the corporate training market. In conjunction with this, EducateOnline reneged on many promises made to provide continual support for

the system. In 2000, members of the university declared their frustration with the system and the senior members of the university started another search for an enterprise-wide learning management system.

Phase Two of Systems Implementation

An evaluation group within the university was created to oversee and evaluate the potential learning management systems. This group contained members from each faculty and service division. Requirements for the new enterprise­wide learning management systems were established, which, according to members interviewed, was one of the least political aspects of the selection process. Each member of the university was given the opportunity to submit their key requirements as to what they would look for in an enterprise-wide learning management system. Teaching staff and students were then given the opportunity to rate each requirement, producing a list of eight key requirements in conjunction with the technological requirements of running on Oracle.

Sixty-four potential learning management systems were identified and this list was quickly reduced down to a working subset of five systems. At one stage of the elimination process, CommunicateOnline was not going to be considered as a serious contender for the enterprise-wide learning management system. This caused problems for the CommunicateOnline advocates. Technological and pedagogical arguments were had between the members of the evaluation group. Eventually, a report from the financial evaluation group was given to the evaluation group, which eliminated CommunicateOnline from further consideration, claiming that CommunicateOnline was not financially viable.

Two potential systems remained: OnlineTeacher and ChalkItUp. Both companies performed demonstrations, highlighting little difference between the two products. OnlineTeacher was deemed by members interviewed as being more user-friendly, but did not allow for flexibility to enhance online teaching and learning. ChalkItUp had more flexibility than OnlineTeacher, but was more difficult to use. The evaluation group finally decided that greater flexibility was more important for future developments in online teaching and learning, and recommended ChalkItUp to the senior members of the university.

Phase Three of Systems Implementation

The implementation of the new enterprise-wide learning management system took place in this phase. When the implementation process was originally planned, it was envisioned that the implementation of ChalkltUp would be enterprise-wide by semester one, 2005. The then Vice-Chancellor retired, and a new Vice-Chancellor was appointed. Shortly after this appointment, the Deputy Vice-Chancellor resigned and a new Deputy Vice-Chancellor was eventually appointed.

Two further changes were made by the new Vice-Chancellor - all units across the university had to have an online presence by semester one 2004, shortening the proposed timeline by twelve months, and providing only six months to train all teaching staff. The second change saw the implementation of a new policy, requiring all units to have either a basic online presence, an extended online presence, or a wholly online presence.

After initial bugs and problems, the faculties migrated their units across from their respective system to ChalkItUp. The Faculty of Commerce and Administration was the most disgruntled faculty, whereas other faculties utilised this opportunity to review their online teaching methods and migrate to ChalkItUp. ChalkltUp has been in operation for four semesters and although has had bugs, has not suffered any major problems. As members of the University of the Eastern Antipodes accept ChalkltUp, further aspects of functionality will be added, strengthening ChalkltUp.

DISCUSSION

The findings of this study demonstrate that the systems selection and implementation process is invariably a complex process that is influenced by factors other than a particular implementation method or tool such as Soft Systems Methodologies (SSM), Structured Analysis, Design and Implementation of Information Systems (STRADIS), Structured Systems Analysis and Design Method (SSADM), Rational Unified Process (RUP), James Martin's RAD (JMRAD), or Effective Technical and Human Interaction of Computer-Based Systems (ETHICS). The story of the selection and implementation process outlined above supports this argument, especially that the systems selection and implementation process is essentially political and non-rational (Mitev, 2001), being influenced by sectional interests, power and factions, often associated with specific stakeholders (Corbitt, 1997).

A two-level understanding of systems selection and implementation merged from the case study. At one level, the systems selection and implementation process went through the process of, planning for the system, developing the

requirements of the system, evaluating each system against the requirements, and selecting and implementing the system. This was a complex and iterative process, affected, even directed, by other levels of organisational influence. The implementation of the system was as much driven by the structures imposed on its use by the administration of the university as it was by the structural limitations of the software and the operating and infrastructure systems supporting it.

The top level of influence was political where direction emerged through statute, policy and the demands of compliance, informed both by discourse and changing power relations. As argued earlier, the focus of previous research on systems implementation has been on the critical success factors involved with implementing a system (Ginzberg, 1981; Rockart, 1979; Delone and McLean, 1992; 2003; Somers and Nelson, 2001) and the actual implementation process (Avison and Fitzgerald, 2003; Davis, 1974). In this study that process was affirmed. However, the nature and form of the system that emerged at the "Full Launch" stage of implementation was fundamentally different from what was proposed in the requirements process. These differences were the inability to personalise the layout and look of the unit area and functionality was limited allowing some users to have a wider variety of tools and features available.

What is significant from this analysis is that the systems process can at the one time follow structured and semi­structured methodologies and at the same time be influenced and in fact changed by the complexities and sectional interest of organisational politics and ensuing struggles for power. This occurs to the extent that an apparently 'normal' systems implementation is at the obvious level a process comprising multiple stages and an iterative form but at the same time delivering content and internal structures which are reflective of social constructed power relations derived from debate over ideology and driven by discourse. The conflicting discourse of academic freedom and managerialism created a series of social dramas over many issues which became the strongest influence on the nature and eventual use of the system itself.

The systems selection and implementation process transforms through the struggles for the exercise of power. This challenges the existing view of systems selection and implementation, that transforming processes are consistent with iteration. In the political realm of organisational power relations, complexity creates chaos. The attempted subjugation of staff is challenged by those actors as a means of political emancipation from control. This search for emancipation by the academic staff was ultimately to enforce structural change in the system itself as management imposed structural parameters on the system which, in effect, restricted its use by academic staff. The selection and implementation of the new enterprise-wide learning management system spanned two and a half years. This time period saw a number of changes and shifts in leadership, resistance and challenges within discourse. This transforming nature of power and politics and the complexity that results in systems selection and implementation is not identified in the systems implementation literature.

This apparent level of influence over the system itself is also couched in lessons learned by the organisation through its history of enterprise-wide selecting and implementing learning management systems. For example, there was a difference in the effectiveness of the implementation between the original enterprise-wide learning management system, EducateOnline, and the new learning management system, ChalkItUp. In the original implementation, there was no policy requiring users to adopt EducateOnline as the then Vice-Chancellor, although interested in providing standardisation across the university, was not driven by a discourse of managerialism. He was driving a discourse of intellectualism, of academic freedom, as a means to improve the research output of the university. Whether users adopted EducateOnline or kept using their current system was not enforced, or required. In the case of the selection and implementation of ChalkItUp, the new Vice-Chancellor did not believe that standardisation and centralisation would occur without policy. Ultimately, ChalkItUp had policy requiring all academic staff to comply with and have a basic online presence for their units. The case study has supported an argument by Somers and Nelson (2001) that the system would not be used across the university had it not been driven from the top.

The story of the selection and implementation process of the enterprise-wide learning management system at the University of the Eastern Antipodes also challenges the accepted view that systems implementation is a linear process, and that in order to "successfully" implement a system, certain steps need to be followed. These steps need to be completed in a linear manner. In other words, we cannot proceed to the next step until the current step has been completed, and that the appropriate outputs have been produced (Avison and Fitzgerald, 2003; Davis, 1974). Rather, this study supports the argument by Nguyen (2000), that the implementation process does not follow this linear method because of the nature of systems implementation and one of the most important elements of systems implementation, the people.

Much of the contestation and debate in the social dramas is reflective at the obvious level to the existence of resistance. Krovi (1993), and Markus (1983) have argued that change is closely related to resistance. What was experienced during the selection and implementation of ChalkltUp was implemented, was in essence, resistance. However, this resistance was only the obvious aspect of the emerging power relations in the organisation. This resistance reflected a deeper discourse over challenges to the discourse of academic authority and intellectualism. Foucault (1978) suggests that power emerges where there is resistance. However, it is not through the front stage issues of power that are of interest, it is through the emerging discourse that the role of power and politics influence the systems selection and implementation process. This case study has highlighted the impact that this can have. However like all case studies the extent to which this level of theorising can be universally applied has to be questioned.

CONCLUSION

Through telling the story of the selection and implementation of an enterprise-wide learning management system at the University of the Eastern Antipodes discourses emerged. The key findings from this study have indicated that the system selection and implementation works at two levels. The low level is the selection and implementation process, which operates for the period of the project. The high level is the arena of power and politics, which runs simultaneously to the selection and implementation process. Challenges for power are acted out in the front stage, or public forums between various actors. The social dramas, as they have been described here, are superfluous to the discourse underpinning the front stage. It is the discourse that remains the same throughout the system selection and implementation process, but it is through various social dramas that reflect those discourses. Furthermore, the enactment of policy legitimises power and establishes the discourse, limiting resistance. Additionally, this research has identified the role of the "State" and its influence at the organisational level, which had been previously suggested in education literature (Ball, 1990).

A set of dramas occurred in the selection process of members of the evaluation group and in the dramas associated with the selection of ChalkltUp. Academic staff created the dramas again because their practice and academic authority were challenged. They argued that the real demand for a new system was not necessary. They also challenged any need for a single system across the university. The university had made its reputation of what it was already doing in teaching and learning. There was no obvious reason to change that. However, the committees were established and a selection process commenced. The academic staff had to recontextualise their practice and accept that power relations about teaching and learning had been, and were again changing. Power was now vested in the selection committee and those advising it. The differentiated set of power relations existing within the university had been usurped by directive and placed, with the authority of the senior members of the university.

The case study indicates that the systems selection and implementation process transforms through the struggles for the exercise of power. This challenges the existing view of systems selection and implementation, that transforming processes are consistent with iteration. In the political realm of organisational power relations, complexity creates chaos. The attempted subjugation of staff is challenged by those actors as a means of political emancipation from control. This search for emancipation by the academic staff was ultimately to enforce structural change in the system itself as management imposed structural parameters on the system which, in effect, restricted its use by academic staff. The selection and implementation of the new enterprise-wide learning management system spanned two and a half years. This time period saw a number of changes and shifts in leadership, resistance and challenges within discourse. This transforming nature of power and politics and the complexity that results in systems selection and implementation is not identified in the systems implementation literature.

The use of social dramas has essentially enabled the researcher to investigate the events in each phase of the selection and implementation process from the front stage perspective, reporting obvious issues in systems implementation, and also from the back stage perspective, identifying the 'hidden' aspects of systems implementation, reflected in the discourse used.

In the front stage (Goffman, 1959) of systems selection and implementation power relations were expressed either openly, or enforced through rules, statutes and policies. Inevitably, the intent of these actions is to establish the role of discipline (Foucault, 1977). In this case study, discipline was imposed through text as policy and enforced by accountability and reporting requirements. This in turn informed the nature of power relations and established the roles of each actor in the systems development and implementation process. Policy, with rules, statutes and decisions reinforced the power status of the decision maker. In tern this can be seen to be an attempt to subjugate the actors. However, it is through the role of discourse, or the practices of talk, text and argument, that continuously form what

which actors speak that creates new discourse and challenges to the prevailing discourse. The challenge to discourse typically emerged as resistance to the new system. Members of the organisation opposed change and attempted to resist the new system through a series of social dramas. It is the recognition of the human factors, and more importantly, the rich view of power and politics in system selection implementation, which is needed to Improve the systems implementation process.

Little research has explored discourse in information systems and therefore, there is little understanding of the role and impact of discourse in information systems research. However, we know from management systems that discourse plays a role in the way organisations operate. It is believed that we need to understand this in information systems. This study has highlighted that discourse does clearly inform process. We need to know more to enrich our understanding. The use of qualitative research with a critical epistemology can provide insight into the discourses involved in information systems. Further system implementation studies will assist in supporting the current research.

REFERENCES Avison, D. E., & Fitzgerald, G. (2003) Information Systems Development: Methodologies, Techniques and Tools (3rd ed.). New York: McGraw-Hill.

Ball, S. J. (1990) Politics and Policy Making in Education: Explorations in Policy Sociology. London: Routledge and Kegan Paul.

Benbasat, 1., Goldstein, D. K., & Mead, M. (1987) The Case Research Strategy in Studies of Information Systems. MIS Quarterly, Il(3), 369-386.

Coad, P., & Yourdon, E. (1991) Object-Oriented Analysis (2nd ed.). New Jersey: Prentice Hall.

Corbitt, B. J. (1995) Homeless Students, Schools and the Policy Process. Unpublished PhD ThesiS, Monash University, Melbourne, Australia.

Corbitt, B. J. (1997) Uncertainty and Equivocality in the Adoption of Electronic Commerce by SMEs' in Australia. Paper presented at the Working Paper, Department ofInformation Systems, University of Melbourne.

Corbitt, B. J., Peszynski, K. J., & Boon, O. (2005) Managing Complexity and Institutionalization: Power and Politics in ERP Implementation in an Australian University. In L. von Hellens, S. Nielsen & J. Beekhuyzen (Eds.), Qualitative Case Studies on Implementation of Enterprise Wide Systems (pp. 160-175). Hershey, PA: Idea Group Publishing.

Cragg, P., & King, M. (1983) Small-firm Computing: Motivators and Inhibitors. MIS Quarterly, 17(1),47-60.

Davis, G. B. (1974) Management Information Systems: Conceptual Foundations, Structure and Development. New York: McGraw-Hill.

DeLone, W. D., & McLean, E. R. (1992) Information Systems Success: The Quest for the Dependent Variable. Information Systems Research, 3(1),60-95.

DeLone, W. D., & McLean, E. R. (2003) The DeLone and McLean Model of Information Systems Success: A Ten-Year Update. Journal of Management Information Systems, 19(4),9-30.

Denscombe, M. (1998) The Good Research Guide: For Small-Scale Social Research Project. Philadelphia: Open University Press.

Downs, E., Clare, P., & Coe, 1. (1988) Structured System Analysis and Design Method: Application and Context (2nd ed.). Hemel Hempstead: Prentice Hall.

Gane, c., & Sarson, T. (1979) Structured Systems Analysis: Tools and Techniques. New Jersey: Prentice Hall.

Ginzberg, M. (1981) Early Diagnosis of MIS Implementation Failure: Promising Results and Unanswered Questions. Management Science, 27(4),457-478.

Goffman, E. (1959) The Presentation of Self in Everyday Life. New York: Doubleday.

Gubrium, J. F., & Holstein, J. A. (2000) Analysing Interpretive Practice. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Qualitative Research (2nd ed., pp. 487-508). California: Sage Publications Inc.

Howarth, D., & Stavrakakis, Y. (2000) Introducing Discourse Theory and Political Analysis. In D. Howarth, A. J. Norval & Y. Stavrakakis (Eds.), Discourse Theory and Political Analysis: Identities, Hegemonies and Social Change (pp. 1-23). Manchester: Manchester University Press.

Fairclough, N. (2001) Critical Discourse Analysis. In A. McHoul & M. Rapley (Eds.), How to Analyse Talk in Institutional Settings: A Casebook of Methods (pp. 25-38). London: Continuum.

Foucault, M. (1977) Discipline and Punish: The Birth of the Prison. London: Penguin Books Ltd.

Foucault, M. (1978) The History of Sexuality, Volume 1: An Introduction (First American Edition ed. Vol. 1). New York: Pantheon Books.

Foucault, M. (1980) Power/Knowledge: Selected Interviews and other Writings, 1972-1977 (C. Gordon, L. Marshall, J. Mepham & K. Soper, Trans.). Brighton, Sussex: Harvester Press.

Krovi, R. (1993) Identifying the Causes of Resistance to IS Implementation: A Change Theory Perspective. Information & Management, 25(6),327-335.

Kvale, S. (1996) Interviews: An Introduction to Qualitative Research Interviewing. Thousand Oaks, Calif.: Sage Publications.

Markus, M. L. (1983) Power, Politics and MIS Implementation. Communications of the ACM, 26(6),430-444.

Mitev, N. N. (2001) The Social Construction of IS Failure: Symmetry, the Sociology of Translation and Politics. In A. Adam, D. Howcroft, H. Richardson & B. Robinson (Eds.), Re­Defining Critical Research in Information Systems (pp. 16-34). Salford: University of Salford.

Nguyen, L. (2000) Incorporating Design Explanation within Formal Object-Oriented Method (FOOM). Unpublished PhD Thesis, Deakin University, Melbourne, Australia.

Ngwenyama, O. K., & Lee, A. S. (1997) Communication Richness in Electronic Mail: Critical Social Theory and the Contextuality of Meaning. MIS Quarterly, 21(2), 145-167.

Orlikowski, W. J. (1992) The Duality of Technology: Rethinking the Concept of Technology in Organizations. Organization Science, 3(3), 398-427.

Rockart, J. F. (1979) Chief Executives Define their Own Data Needs. Harvard Business Review, 57(2),81-93.

Royall, 1. (2002, 2nd Feb) RMIT Online Disaster. Herald Sun, p. 11.

Sauer, C. (1993) Why Information Systems Fail: A Case Study Approach. Henley-on-Thames: Alfred Waller.

Somers, M. T., & Nelson, K. (2001) The Impact of Critical Success Factors across the Stages of Enterprise Resource Planning Implementations. Paper presented at the Proceedings of the 34th Hawaii International Conference on Systems Sciences, Hawaii.

Turner, V. W. (1957) Schism and Continuity in an African society - A study of Ndembu village life. New York: Humanities Press, inc.

Turner, V. W. (1980) Social Dramas and Stories about Them. Critical Inquiry, 7, 141-168.

COPYRIGHT

Peszynski, K. J., Corbitt, B. J. and Coulthard, D. W. © 2005. The authorls assign Griffith University a non-exclusive licence to use this document for personal use provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Griffith University to publish this document in full in the Conference Proceedings. Such documents may be published on the World Wide Web, CD-ROM, in printed form, and on mirror sites on the World Wide Web. Any other usage is prohibited without the express permission of the authors.